U.S. patent number 7,688,627 [Application Number 11/861,102] was granted by the patent office on 2010-03-30 for flash memory array of floating gate-based non-volatile memory cells.
This patent grant is currently assigned to Intersil Americas Inc.. Invention is credited to Michael D. Church, Hosam Haggag, Alexander Kalnitsky, Edgardo Laber, Prabhjot Singh.
United States Patent |
7,688,627 |
Haggag , et al. |
March 30, 2010 |
Flash memory array of floating gate-based non-volatile memory
cells
Abstract
A flash memory array comprises a plurality of memory cells
organized in a matrix of rows and columns. Each of the memory cells
includes a floating gate memory transistor having a source region
and a drain region, and a coupling capacitor electrically connected
to the memory transistor. A plurality of word lines are each
electrically connected to the capacitor in each of the memory cells
in a respective row. A first set of bit lines are each electrically
connected to the drain region of the memory transistor in each of
the memory cells in a respective column. A plurality of high
voltage access transistors are each electrically connected to a bit
line in the first set of bit lines. A second set of bit lines are
each electrically connected to the source region of the memory
transistor in each of the memory cells in a respective column.
Various combinations of voltages can be applied to the word lines
and the first and second sets of bit lines in operations to erase,
program, inhibit, or read the logic state stored by the memory
transistor in one or more of the memory cells.
Inventors: |
Haggag; Hosam (Mountain View,
CA), Kalnitsky; Alexander (San Francisco, CA), Laber;
Edgardo (San Jose, CA), Singh; Prabhjot (San Jose,
CA), Church; Michael D. (Sebastian, FL) |
Assignee: |
Intersil Americas Inc.
(Milipitas, CA)
|
Family
ID: |
39886783 |
Appl.
No.: |
11/861,102 |
Filed: |
September 25, 2007 |
Prior Publication Data
|
|
|
|
Document
Identifier |
Publication Date |
|
US 20080266958 A1 |
Oct 30, 2008 |
|
Related U.S. Patent Documents
|
|
|
|
|
|
|
Application
Number |
Filing Date |
Patent Number |
Issue Date |
|
|
60913586 |
Apr 24, 2007 |
|
|
|
|
Current U.S.
Class: |
365/185.08;
365/185.1; 365/185.06; 365/185.05 |
Current CPC
Class: |
G11C
16/0416 (20130101) |
Current International
Class: |
G11C
14/00 (20060101); G11C 16/04 (20060101) |
Field of
Search: |
;365/185.08 |
References Cited
[Referenced By]
U.S. Patent Documents
Primary Examiner: Mai; Son L
Attorney, Agent or Firm: Fogg & Powers LLC
Parent Case Text
This application claims the benefit of priority to U.S. Provisional
Application No. 60/913,586, filed on Apr. 24, 2007, the disclosure
of which is herein incorporated by reference.
Claims
What is claimed is:
1. A flash memory array comprising: a plurality of memory cells
organized in a matrix of rows and columns, each of the memory cells
comprising: a floating gate memory transistor having a source
region and a drain region; and a coupling capacitor electrically
connected to the memory transistor; a plurality of word lines each
electrically connected to the coupling capacitor in each of the
memory cells in a respective row; a first set of bit lines each
electrically connected to the drain region of the memory transistor
in each of the memory cells in a respective column; a plurality of
high voltage access transistors each electrically connected to a
bit line in the first set of bit lines; and a second set of bit
lines each electrically connected to the source region of the
memory transistor in each of the memory cells in a respective
column; wherein various combinations of voltages can be applied to
the word lines and the first and second sets of bit lines in
operations to erase, program, read, or inhibit a logic state stored
by the memory transistor in one or more of the memory cells.
2. The flash memory array of claim 1, wherein the memory cells are
organized in a two-dimensional array.
3. The flash memory array of claim 1, wherein the word lines each
provide a bus line that is electrically connected to each capacitor
in a respective row.
4. The flash memory array of claim 1, wherein the access
transistors are configured to withstand a high voltage on at least
a drain terminal thereof.
5. The flash memory array of claim 1, wherein each of the access
transistors are electrically connected to respective memory
transistors in the same column.
6. The flash memory array of claim 1, wherein the second set of bit
lines each provide a bus line that is electrically connected to
each memory transistor in a respective column.
7. The flash memory array of claim 1, wherein the memory cells each
comprise a capacitor terminal electrically connected to the
coupling capacitor, and a program terminal electrically connected
to the memory transistor.
8. The flash memory array of claim 1, wherein the erase and program
operations are carried out by Fowler-Nordheim tunneling.
9. A method of erasing a single bit in a memory cell of a flash
memory array according to claim 1, the method comprising: grounding
a word line connected to the coupling capacitor in the memory cell
to be erased; applying a first voltage to a bit line connected to
the source region of the memory transistor in the memory cell to be
erased; applying a second voltage to word lines not connected to
the memory cell to be erased, the second voltage being about
one-half of the first voltage; grounding a bit line connected to
the drain region of the memory transistor in the memory cell to be
erased; and applying a third voltage to bit lines, in the second
set of bit lines, not connected to the memory cell to be erased,
the third voltage being about one-half of the first voltage.
10. The method of claim 9, wherein erasing of the single bit is
carried out by Fowler-Nordheim tunneling.
11. A method of programming a single bit in a memory cell of a
flash memory array according to claim 1, the method comprising:
applying a first voltage to a word line connected to the coupling
capacitor in the memory cell to be programmed; grounding a bit line
connected to the source region of the memory transistor in the
memory cell to be programmed; applying a second voltage to bit
lines, in the second set of bit lines, not connected to the memory
cell to be programmed, the second voltage being about one-half of
the first voltage; grounding a bit line connected to the drain
region of the memory transistor in the memory cell to be
programmed; and grounding word lines not connected to the memory
cell to be programmed.
12. The method of claim 11, wherein programming the single bit is
carried out by Fowler-Nordheim tunneling.
13. A method of reading a single bit in a memory cell of a flash
memory array according to claim 1, the method comprising: grounding
a word line connected to the coupling capacitor in the memory cell
to be read; applying a first voltage to a bit line connected to the
source region of the memory transistor in the memory cell to be
read; applying a second voltage to a bit line connected to the
drain region of the memory transistor in the memory cell to be
read, the second voltage being larger than the first voltage or
enough to turn on an access transistor coupled to the bit line
connected to the drain region; and applying a third voltage to word
lines not connected to the memory cell to be read, the third
voltage being at least about half the first voltage.
14. The method of claim 13, wherein the first voltage is about 1
volt, and the second voltage is about 5 volts.
15. The method of claim 14, wherein the third voltage is about 2-5
volts.
Description
BACKGROUND
Non-volatile memory devices are used in applications that require
stored information to be retained even when the memory devices are
not powered. Examples of non-volatile memory devices include
read-only memory (ROM), magnetic computer storage devices (e.g.,
hard disks, floppy disk drives), optical disc drives, erasable
programmable read-only memory (EPROM), electrically erasable
programmable read-only memory (EEPROM), and flash memory.
EEPROMs have been widely used in electronic products because of
their non-volatile functions of electrically writing and erasing
data. An EEPROM can be programmed and erased electrically using
field emission (also known as "Fowler-Nordheim tunneling"). In the
erase method of a conventional EEPROM, stored electric charges will
move from a floating gate and be removed due to the Fowler-Nordheim
tunneling (F-N tunneling) effect.
Flash memory is like EEPROM in that it can also be electrically
erased and programmed. Flash memory stores information in an array
of floating gate transistors, called "cells," each of which
traditionally stores one bit of information. Newer flash memory
devices, sometimes referred to as multi-level cell devices, can
store more than 1 bit per cell by using more than two levels of
electrical charge placed on the floating gate of a cell.
A limitation of conventional flash memory is that although it can
be read or programmed a byte or a word at a time in a random access
fashion, the flash memory has to erase multiple memory locations
("block") at a time. This generally sets all bits in the block to
1. Changing a single byte is only possible by rewriting a whole
block. Starting with a freshly erased block, any location within
that block can be programmed. However, once a bit has been set to
0, only by erasing the entire block can it be changed back to 1. In
other words, conventional flash memory offers random-access read
and programming operations, but cannot offer arbitrary
random-access rewrite or erase operations.
Another limitation of conventional flash memory as applied to
standard complementary metal oxide semiconductor (CMOS) processes
is the number of additional masks required to embed a flash cell
and the peripheral circuitry used in generation and routing of high
voltage signals.
SUMMARY
The present invention is related to a flash memory array comprising
a plurality of memory cells organized in a matrix of rows and
columns. Each of the memory cells comprises a floating gate memory
transistor having a source region and a drain region, and a
coupling capacitor electrically connected to the memory transistor.
A plurality of word lines are each electrically connected to the
coupling capacitor in each of the memory cells in a respective row.
A first set of bit lines are each electrically connected to the
drain region of the memory transistor in each of the memory cells
in a respective column. A plurality of high voltage access
transistors are each electrically connected to a bit line in the
first set of bit lines. A second set of bit lines are each
electrically connected to the source region of the memory
transistor in each of the memory cells in a respective column.
Various combinations of voltages can be applied to the word lines
and the first and second sets of bit lines in operations to erase,
program, read, or inhibit a logic state stored by the memory
transistor in one or more of the memory cells.
BRIEF DESCRIPTION OF THE DRAWINGS
Features of the present invention will become apparent to those
skilled in the art from the following description with reference to
the drawings. Understanding that the drawings depict only typical
embodiments of the invention and are not therefore to be considered
limiting in scope, the invention will be described with additional
specificity and detail through the use of the accompanying
drawings, in which:
FIG. 1 is a schematic cross section view of a multiple time
programmable memory cell according to one embodiment;
FIG. 2 is schematic representation of a flash memory array
according to one embodiment, which illustrates the operation of an
erase function;
FIG. 3 depicts the flash memory array of FIG. 2, but illustrates
the operation of a programming function; and
FIG. 4 depicts the flash memory array of FIG. 2, but illustrates
the operation of a read function.
DETAILED DESCRIPTION
In the following detailed description, embodiments are described in
sufficient detail to enable those skilled in the art to practice
the invention. It is to be understood that other embodiments may be
utilized without departing from the scope of the present invention.
The following detailed description is, therefore, not to be taken
in a limiting sense.
The complementary metal oxide semiconductor (CMOS) fabrication
technology is a common fabrication technique for integrated
circuits such as a p-channel and an n-channel MOSFET (metal oxide
semiconductor field-effect transistor) referred to herein as PMOS
and NMOS, respectively. Since CMOS technology continues to use 5V
input-output (I/O) devices with a gate dielectric thickness in the
range of about 10-15 nm, it is necessary to operate a memory cell
at high positive erase voltages (e.g., about 12 V to about 22
V).
The present invention is directed to a two transistor (2T) flash
memory array with EEPROM emulation capability that is built with
PMOS floating gate-based non-volatile memory cells. The present
flash memory array can be fabricated using conventional CMOS
technology with a thick gate oxide (e.g., about 125 .ANG.). The
memory cells in the flash memory array can tolerate relatively high
erase voltages.
The present invention addresses the limitations of conventional
flash memory by providing a memory cell built out of elements
native to CMOS technology. High voltage routing can be accomplished
by devices capable of sustaining required voltages on one terminal
(drain) only, such as extended drain or drift metal oxide
semiconductor (DMOS) devices or true high voltage (HV) CMOS devices
if present in a given process.
The present flash memory array is constructed with a two transistor
architecture in each memory cell, including a storage transistor
and a coupling capacitor in each cell. An access (or pull down)
transistor is shared by a plurality of cells attached to a common
column (bit line). Both the program and the erase operations are
carried out by Fowler-Nordheim tunneling. The present flash memory
array can be operated in a mode similar to traditional EEPROM in
that the flash memory array allows byte by byte (and down to bit
level) programming and erase.
Further details of the present invention are described as follows
with respect to the drawings.
FIG. 1 depicts a cross section of a multiple time programmable
(MTP) PMOS floating-gate based non-volatile memory cell 100 that
can be employed in the flash memory array of the present invention.
All elements of memory cell 100 can be built out of components
intrinsic to a generic CMOS process flow with thick I/O oxide
devices. The memory cell 100 generally includes a floating gate
memory transistor 120, a coupling capacitor 140, and a high voltage
access transistor 160. Each of these components is discussed in
further detail hereafter.
The memory transistor 120 includes a p-type source region 122
spaced apart from a p-type drain region 124, both of which are
formed in an n-type well 121 (N-well). The n-type well 121 is in
turn formed in a p-type substrate 102. A channel region 126 is
defined between source region 122 and drain region 124. A layer of
gate oxide 128 is formed over channel region 126, and a floating
gate 130 such as a layer of polysilicon is formed over gate oxide
128. Since gate 130 is isolated, it is often referred to as a
floating gate. An n-type region 123 is also formed within n-type
well 121 adjacent to p-type source region 122. The n-type region
123 provides an ohmic body tie to n-type well 121, so n-type well
121 can be tied to a terminal. A contact region 132 such as a layer
of silicide is formed over p-type source region 122 and n-type
region 123. A layer of silicide 134 can also be formed over p-type
drain region 124. Silicide formation may be optionally excluded
from the floating gate area by application of a conventional
silicide block process. A field oxide (FOX) region 136 isolates
memory transistor 120 from high voltage access transistor 160.
In the embodiment shown in FIG. 1, memory transistor 120 is a
floating gate PMOS transistor. The memory transistor 120 may also
be referred to as a storage transistor, because it can be
programmed to store a logic state.
The coupling capacitor 140 includes spaced apart n-type diffusion
regions 143 and 145 formed within an n-type well (N-well) 141,
which is formed in p-type substrate 102. A channel region 146 is
defined between n-type diffusion regions 143 and 145, and a layer
of gate oxide 148 is formed over channel region 146. A polysilicon
top plate gate layer 150 is formed over gate oxide 148 to form the
coupling capacitor top plate. A contact region 153 such as a layer
of silicide is formed over n-type diffusion region 143, and a layer
of silicide 155 is formed over n-type diffusion region 145. A FOX
region 135 isolates capacitor 140 from memory transistor 120.
Alternatively, the coupling capacitor can be formed as a
conventional PMOS transistor or a gated diode capacitor (not
shown).
The high voltage access transistor 160 includes both a p-type well
161 (P-well) and an n-type well 171 (N-well). The n-type well 171
forms the drain of transistor 160, with an n-type region 174
providing an ohmic body tie to a silicide contact region 176. An
n-type source region 162 is formed in p-type well 161. A channel
region 166 is defined between n-type source region 162 and n-type
well 171. A layer of gate oxide 168 is formed over part of the
P-well and part of the N-well (including where they abut one
another), and a gate 180 is formed over gate oxide 168, resulting
in gate 180 being over channel 166. A p-type region 163 is also
formed within p-type well 161, next to n-type source region 162.
The p-type region 163 provides an ohmic body tie to p-type well 161
so that p-type well 161 is tied to ground.
A layer of silicide 165 is formed over n-type source region 162 and
p-type region 163, to thereby form a contact region that is shown
as being connected to ground. A FOX region 178 is formed in n-type
well 171 to isolate silicide contact region 176 from gate 180. It
is this isolation that enables access transistor 160 to withstand
the higher voltages which occur during an erase operation.
In the embodiment shown in FIG. 1, high voltage access transistor
160 is a asymmetrical high voltage n-channel MOSFET (NDMOS) device.
The high voltage access transistor 160 may also be referred to as a
high voltage select transistor. Because of its extended drain 171,
access transistor 160 can also be referred to as a high voltage
extended drain NMOS transistor. Alternatively, an NDMOS device
designed for optimum performance of a Bipolar CMOS DMOS (BCD)
process can be employed as an access device. In a further
alternative embodiment, if a technology supports a set of high
voltage devices, these can be used as access transistors.
A symmetrical parasitic NDMOS device can also be used as an access
transistor. This device will have two regions of field oxide under
both edges of the gate polysilicon, with two N-wells used as S/D
extensions and a P-well region entirely under the gate polysilicon
region of the DMOS device (structure not shown).
A p-type well (P-well) 192 is used to electrically isolate n-type
well 141 from n-type well 121, without contacting either of n-type
well 121 or n-type well 141. Similarly, a p-type well (P-well) 194
can electrically isolate n-type well 121 from n-type well 171,
without contacting either of n-type well 121 or n-type well 171.
The inclusion of p-type wells 192, 194 separated from N-wells by a
region of lightly doped Si allows for application of higher
voltages by preventing punch-thru between the neighboring n-type
wells 121, 141 and the neighboring n-type wells 121, 171. Because
of their function, p-type wells 192, 194 can be referred to as
isolation wells. A similar approach can be used to increase the
breakdown voltage of the symmetrical or the asymmetrical NDMOS
(i.e., P-well 161 can be separated from N-well 171 by a region of
lightly doped Si).
In accordance with some embodiments of the present invention, each
gate oxide layer 128, 148 and 168 can have a gate oxide thickness
that is the same as the gate oxide thickness of CMOS devices that
are used as I/O interface devices having an operating voltage of
5V. In other words, the thickness for gate oxide layers 128, 148
and 168 can be native to the fabrication process for 5V I/O
devices. This enables capacitor 140 and transistors 120, 160 to be
made using standard CMOS processes. For example, each gate oxide
layer 128, 148 and 168 can have a thickness in the range of about
10-15 nm (i.e., about 100-150 .ANG.). In some embodiments, the
thickness of each gate oxide layer 128, 148 and 168 can be at least
about 12 nm (i.e., at least about 120 .ANG.). In other embodiments,
a gate oxide layer thickness of up to about 20 nm (i.e., up to
about 200 .ANG.) can be used, enabling such embodiments to be
useful with devices having even higher I/O voltages.
The memory cell 100 includes three terminals. A capacitor terminal
(Vcap) is electrically connected to contact region 153 of capacitor
140. A program terminal (Vpp) is electrically connected to contact
region 132 of memory transistor 120. The Vpp terminal can be used
for both programming and erasing the memory cell. A control
terminal (Vcg), also referred to as a select or access terminal, is
electrically connected to gate 180 of access transistor 160.
The top plate 150 of capacitor 140 is electrically connected to
floating gate 130 of memory transistor 120 by a trace 137. There is
no contact to floating gate 130 of memory transistor 120. The
capacitor 140 is designed to have a much higher capacitance (e.g.,
about 10 times) than that of memory transistor 120. The capacitor
140 capacitively couples floating gate 130 to n-type well 141.
Thus, when a high erase voltage (e.g., about 14-22 V) is applied to
the Vpp terminal, a voltage drop across gate oxide 128 forms a
sufficient electric field for Fowler-Nordheim tunneling such that
electrons are tunneled off floating gate 130. As also shown in FIG.
1, p-type drain region 124 of memory transistor 120 is electrically
connected to n-type drain region 171 (through n-type region 174 )
of access transistor 160 by a trace 138. Use of the NDMOS access
transistor allows high voltages required for cell erase to be
applied to the Vpp terminal without causing two junctions
(N-well-P+_in the memory device in series with the N-well-P-well or
N-well to substrate) breakdown.
Further details with respect to memory cell 100, including
additional alternative embodiments, are described in copending U.S.
application Ser. No. 11/498,672, filed on Aug. 2, 2006, the
disclosure of which is incorporated herein by reference.
FIG. 2 is a schematic representation of a two-dimensional 2 T flash
memory array 200 according to one embodiment of the invention. The
flash memory array 200 includes a plurality of memory cells 210
organized in a matrix of rows and columns. Although three rows and
three columns are shown, it should be understood that any number of
rows and columns of memory cells can be formed, and is only limited
by the number of cells that can be formed on a particular chip.
Each of the memory cells 210 includes a floating gate memory
transistor 220 and a coupling capacitor 240, which can correspond
to memory transistor 120 and capacitor 140, respectively, as
discussed with respect to FIG. 1. The floating gate memory
transistor 220 is configured to store a logic state. The coupling
capacitor 240 is electrically connected to floating gate memory
transistor 220.
The memory cells 210 are connected together in a matrix by word
lines and bit lines arranged in a plurality of rows and columns. A
plurality of word lines (WL1, WL2, WL3, etc.) provides a Vcap bus
line that is electrically connected to each capacitor 240 in a
respective row. A first set of bit lines (BL1, BL2, BL3, etc.) are
each electrically connected to a drain region of memory transistor
220 in each of the memory cells in a respective column. A plurality
of high voltage access transistors 260, which are configured to
withstand a high voltage on at least a drain terminal, are each
electrically connected to a respective bit line BL1, BL2, BL3, etc.
Each of these bit lines provides a Vcg select signal that
electrically connects an access transistor 260 to a respective
memory transistor 220 connected to the first set of bit lines in
each column.
A second set of bit lines (BL1', BL2', BL3', etc.) each provides a
bus line that is electrically connected to a source region of
memory transistor 220 in each of the memory cells along a
respective column. Various combinations of voltages can be applied
to the word lines and the bit lines to program, inhibit, read, or
erase the logic state stored by memory transistor 220 in one or
more of the memory cells.
In a method of erasing a single bit in a memory cell of flash
memory array 200, a word line connected to capacitor 240 in the
memory cell to be erased is grounded. A first voltage is applied to
a bit line connected to the source region of memory transistor 220
in the memory cell to be erased. A second voltage is applied to
word lines not connected to the memory cell to be erased, with the
second voltage being about one-half of the first voltage. A bit
line connected to the drain region of memory transistor 220 in the
memory cell to be erased is also grounded. A third voltage is
applied to bit lines, in the second set of bit lines, not connected
to the memory cell to be erased, with the third voltage being about
one-half of the first voltage.
In a method of programming a single bit in a memory cell of flash
memory array 200, a first voltage is applied to a word line
connected to capacitor 240 in the memory cell to be programmed. A
bit line connected to the source region of memory transistor 220 in
the memory cell to be programmed is grounded. A second voltage is
applied to bit lines, in the second set of bit lines, not connected
to the memory cell to be programmed, with the second voltage being
about one-half of the first voltage. A bit line connected to the
drain region of memory transistor 220 in the memory cell to be
programmed is grounded. Word lines that are not connected to the
memory cell to be programmed are also grounded.
In a method of reading a single bit in a memory cell of flash
memory array 200, a word line connected to capacitor 240 of the
memory cell to be read is grounded. A first voltage is applied to a
bit line connected to the source region of memory transistor 220 in
the memory cell to be read. A second voltage is applied to a bit
line connected to the drain region of memory transistor 220 in the
memory cell to be read. The second voltage is applied to the gate
of access transistor 260 at terminal Vcg, with the access
transistor coupled to the bit line connected to the drain region.
The second voltage is larger than the first voltage or enough to
turn "on" the access transistor. A third voltage is applied to word
lines not connected to the memory cell to be read, with the third
voltage being at least about half the first voltage.
Further details regarding erase, program, and read operations for
the present flash memory array are discussed hereafter. As
mentioned previously, erase and program operations in an EEPROM
memory array are carried out by Fowler-Nordheim (F-N) tunneling.
The use of F-N tunneling for both programming and erasing a memory
cell in the EEPROM memory array requires the lowest possible
operating current. Such an operation mode can also increase cell
endurance and data retention time. FIG. 2 further illustrates the
operation of an F-N erase function at the bit level for a cell M of
flash memory array 200. In order to erase cell M, an erase voltage
sufficient to create an electric field needed for F-N tunneling
through the gate oxide of the memory transistor 220 is applied to
the Vpp terminal of cell M, while the Vcg terminal and the Vcap
terminal of cell M are connected to ground. For example: (1) To
erase cell M at WL=2 and BL=2: WL2=0, BL2'=Vpp, Vcg=0; (2) Column 2
inhibit: for all cells on BL2' apply Vpp/2 on WL1 and WL3. (3) All
cells on WL1 and WL3 will see Vpp/2 thus: a. If a cell is already
erased, Vpp/2 is not enough to program it. b. If a cell is already
programmed, Vpp/2 will not disturb its state. (4) Row inhibit: all
cells on WL2 will see VWL=0; a. If a cell on WL2 is erased, VWL2=0
will cause no disturb; b. Same if a cell on WL2 is programmed.
In order to erase cell M (one bit) of memory array 200, capacitor
240 in cell M needs to go high (e.g., about 17-18 V). Accordingly,
bit line BL2 needs to go high and word line WL2 needs to go to a
zero potential. With this configuration, a voltage is applied to
bit line BL2' (VPP), and this voltage is divided between memory
transistor 220 and capacitor 240 in cell M. Since the memory
transistor is designed to be considerably smaller than the
capacitor, most of the voltage is dropped across the memory
transistor. The potential is positive on the N-well and zero or
close to zero on the floating gate of the memory transistor. By
applying these potentials, electrons can be injected from the
floating gate into the substrate of cell M.
In order to protect the rest of memory array 200 from being erased,
word lines that are not grounded such as WL1 and WL3, which are
connected to cells in the same column as cell M, need to be kept at
approximately one-half the potential VPP to prevent an erase. In
order to prevent erasure of cells to the left and to the right of
cell M, bit lines BL1' and BL3' also need to be kept at about
one-half VPP.
In addition, the erase operation can also be used to simultaneously
erase all cells connected to a common Vpp bus. If desired, the
whole flash memory array can be erased at the same time by applying
high voltage to every bit line prime (BL1', BL2', BL3', etc.) and
ground to every word line (WL1, WL2, WL3, etc.).
FIG. 3 depicts flash memory array 200 as described above with
respect to FIG. 2, but illustrates the operation of an F-N
programming function at the bit level. The use of F-N programming
provides the lowest power consumption available to program a memory
cell. In general, a programming voltage level is achieved by
application of a high voltage to the Vcap terminal, while the Vcg
terminal and the Vpp terminal are grounded. In order to program a
single cell, a high voltage is applied to an associated word
line.
A program condition for the programming function of flash memory
array 200 at the bit level for a single cell M is as follows: (1)
To program cell M at WL=2 and BL=2: WL2=Vpp, BL2'=0, Vcg=0; (2)
Row-2 inhibit: all cells on WL2 will see Vpp; apply Vpp/2 on BL1'
and BL3'. a. If a programmed cell is on BL1' or BL3' sees Vpp/2;
not enough to erase, b. If an erased cell is on BL1' or BL3' sees
Vpp/2; don't care. (3) Column 2 inhibit: all cells on BL2' will see
BL=0, WL1=WL3=0 or floating. No disturb in WL1 or WL2 for erased or
programmed cells.
In a programming mechanism to program cell M (one bit), a high
voltage is applied to word line WL2 (VPP). When the high VPP is
applied to capacitor 240 of cell M and ground is applied to bit
line BL2', the inverse of the erase operation occurs. A high
voltage is dropped across the floating gate of memory transistor
220 of cell M, with zero or close to zero volts in the substrate of
cell M, which causes injection of electrons from the substrate into
the floating gate.
One cell can be programmed one bit at a time by keeping other cells
from being programmed. For example, with respect to the cells to
the left and to the right of cell M in FIG. 3, if BL1' and BL3'
both have a voltage applied which is approximately one-half of VPP,
the programming of these cells is inhibited. For cells on different
word lines, it is enough to ground these word lines and no
programming will take place.
FIG. 4 depicts flash memory array 200 as described above with
respect to FIG. 2, but illustrates the operation of a read function
at the bit level for a cell M. The memory cell M can be read by
applying a significantly lower read voltage (e.g., approximately 1
V) to the Vpp terminal while the Vcg terminal receives a select
voltage level such as about 5 V.
A program condition for the read function of flash memory array 200
at the bit level for cell M is as follows: (1) To read cell M at
WL=2 and BL=2: VBL2'=1V, Vcg of BL2=5V; (2) Column inhibit: all
cell on BL2' will see 1V; a. If a cell is already programmed on BL2
(it is on), then to inhibit read disturb, VWL1=VWL3=VPP/2 or at
least the logic level (e.g., about 5V), i. Then, all cells on WL1
and WL3 will see VWL inhibit. For an erased cell, VPP/2 is not
enough to program. ii. For a programmed cell (on), don't care. b.
If a cell is already erased on BL2'=1V, don't care.
In a read operation for cell M (one bit), which is attached to bit
line BL2, the cell is activated by applying a relatively low
voltage (about 1V) to memory transistor 220 of the cell. If cell M
has been previously programmed such that it is in a conductive
state, the cell will conduct word line voltage WL2=0, and will have
a current flow from bit line BL2' through the cell, and through the
column access transistor 260 (VCG) connected to bit line BL2. When
1 volt is applied to bit line BL2' for a read operation, the cells
above and below cell M in the same column may conduct and such
conduction needs to be blocked. Accordingly, the word lines WL1,
WL3, etc. for the cells in the same column as cell M that are not
being read need to be at a high voltage (e.g., VPP of about 2-5
volts) in order to block conduction. If any of these unread cells
are programmed, a higher voltage level is needed to block
conduction than if the cells have been erased.
The present invention may be embodied in other specific forms
without departing from its essential characteristics. The described
embodiments and methods are to be considered in all respects only
as illustrative and not restrictive. The scope of the invention is
therefore indicated by the appended claims rather than by the
foregoing description. All changes that come within the meaning and
range of equivalency of the claims are to be embraced within their
scope.
* * * * *